Physics-Scene Visual Reasoning on Physics
54.35AccuracyS1-VL-32B-RL
Evaluation results
| Method | Links | |
|---|---|---|
| S1-VL-32B-RLParameter Count=32B, Training Phase=RL2026.04 | 54.35 | |
| GPT-52026.04 | 48.34 | |
| Qwen3-VL-235B-A22B-ThinkingParameter Count=235B-A22B, Reasoning Mode=Thinking2026.04 | 46.03 | |
| Intern-S1Parameter Count=235B+6B2026.04 | 44.95 | |
| S1-VL-32B-SFTParameter Count=32B, Training Phase=SFT2026.04 | 43.8 | |
| Qwen3-VL-32B-ThinkingParameter Count=32B, Reasoning Mode=Thinking2026.04 | 41.71 | |
| Gemini 2.5 Pro2026.04 | 40 | |
| Gemini 2.5 Flash2026.04 | 37.1 | |
| Intern-S1-miniParameter Count=8B2026.04 | 30.53 | |
| Thyme-VLParameter Count=7B2026.04 | 12.41 |